It is highly desirable to be able to probe biological activities with subcellular resolution deep inside live organisms. To this end, light microscopy is the most powerful and versatile modality. In particular, by employing a spatially confined excitation via a nonlinear transition, two-photon excited fluorescence microscopy has become indispensable for imaging scattering samples such as brain. However, as the incident laser power drops exponentially with imaging depth due to scattering loss, the out-of-focus fluorescence eventually overwhelms the in-focal signal. The resulting loss of imaging contrast, S/B, defines a fundamental imaging-depth limit (about 1 mm for mouse brain tissues), which cannot be overcome by increasing excitation intensity. Thus, how to image deeper than the fundamental imaging- depth limit poses a grand challenge for many biomedical studies including neuroscience, embryology and oncology. Novel optical imaging techniques that accomplish this goal would undoubtedly open up new avenues, transforming our ability to monitor living systems. We propose to address this challenge by exploring a unique probe-centered strategy as opposed to the popular wave-centered approaches. We realize that these exists a special class of imaging probes that can occupy metastable on- and off- states which can be manipulated by external light at proper wavelengths. By harnessing this unique class of photoswitchable probes, we propose to develop a novel platform of super-nonlinear (higher than quadratic dependence of its signal on the laser intensity) fluorescence microscopy which should be able to promote the S/B contrast and extend the fundamental imaging-depth limit of two-photon microscopy. Specifically, we will focus on two seemingly opposite but actually related techniques which couple photo-switchable probes with two-photon microscopy, namely, multiphoton activation and imaging (MPAI) and multiphoton deactivation and imaging (MPDI). Our preliminary results have demonstrated the validity of both MPAI and MPDI on three-dimensional tissue phantoms. Moreover, the 4th order super-nonlinear dependence of the signal on laser intensity was also verified experimentally. Hence, we aim to further develop and perfect the technique to the stage where it can be applied to imaging various scattering samples, particularly brain tissues, with a much better S/B contrast and depth penetration. Specifically, we plan to (1) systematically evaluate the emerging generation of photoswitchable probes (including fluorescent proteins and synthetic dyes); (2) apply the most promising probes into brain tissue slice imaging and, (3) ultimately be able to perform in vivo deep brain MPAI or MPDI with 2.4 times deeper than what two-photon microscopy can ever achieve. The proposed technical innovation has the potential to change the future landscape of in vivo light microscopy, take bio-imaging into new areas of biomedicine that have been previously uncharted.